Transforming growth factor beta1 (TGFbeta1) is a potent growth inhibitor of epithelial cells; however, its role in carcinogenesis has not been clearly defined. Previous studies have suggested that TGFbeta1 plays seemingly contradictory roles in carcinogenesis: suppressing tumor growth at earlier stages, but promoting malignant progression at later stages. This proposal will investigate in vivo mechanisms underlying the functional switch of TGFbeta1 in skin carcinogenesis. To achieve this, transgenic mice have been generated have been generated in which a TGFbeta1 transgene can be focally induced in the epidermis by total application of a sub-pharmacological dose of a progesterone antagonist (e.g., ZK98.734). TGFbeta1 transgenic mice will be exposed to a chemical carcinogenesis protocol that allows tumorigenesis to develop in the discrete stages of initiation, promoting and malignant conversion. By applying ZK98.734 to a focal of the skin or to tumors at different stages, TGFbeta1 transgene expression can be induced in situ, thus, stage-specific roles of TGFbeta1 over-expression in carcinogenesis can be assessed. Additionally, to determine the consequence of blocking TGFbeta signaling in skin carcinogenesis, transgenic mice expressing a dominant negative TGFbeta type II receptor (deltabetaRII) in the epidermis have been generated, and will be tested for susceptibility to chemical carcinogenesis. Downstream targets of TGFbeta1 involved in multistage carcinogenesis will also be identified from these two complementary mouse models. To assess whether TGFbeta1 induces tumor progression via effects on the stroma, TGFbeta1 mice will be matured with the deltabetaRII mice. Because the deltabetaRII is only expressed in the epidermis and blocks TGFbeta1-induced growth inhibition in keratinocytes. TGFbeta1 will exert its paracrine effect n the dermis where cells still possess functional TGFbeta receptors. Once chemically-induced non-invasive tumors arise in these mice, ZK98.734 will be applied to tumors to observe whether TGFbeta1 induction accelerates tumor invasion and metastasis. Although not a specific aim in this proposal, transgenic mice that develop TGFbeta- resistant tumors, which have an invasive and metastatic phenotype, will provide useful models for testing therapeutic approaches.